The influence of environmental calcium concentrations on calcium flux, compensatory drinking and epithelial calcium channel expression in a freshwater cartilaginous fish

The regulation and balance of ionic calcium (Ca²⁺) is vital to all organisms. Ca²⁺ is necessary for a multitude of physiological processes and is important as a structural element in both invertebrates and vertebrates. Aquatic vertebrates have a source of ambient Ca²⁺ in water, unlike terrestrial vertebrates, which can only uptake Ca²⁺via ingestion (Flik et al., 1995). Fishes, as aquatic vertebrates, have either a bony skeleton or a cartilaginous skeleton. While bony fishes are abundant in both freshwater and seawater environments, cartilaginous fishes generally occupy high-calcium marine environments, and examples of cartilaginous fishes inhabiting freshwater environments are comparatively rare (Ballantyne and Robinson, 2010; Compagno, 1990). Interestingly, many of the fishes with this body design and life history are primitive fishes, which are also known to maintain very low circulating concentrations of Ca²⁺ (Allen et al., 2009c).

Fishes are able to maintain plasma Ca²⁺ concentrations within fairly narrow ranges, although the ranges are generally not as narrow as in terrestrial vertebrates (Dacke, 1979; Flik and Verbost, 1993). Internal Ca²⁺ concentrations are maintained by a balance between influx and efflux. However, the influence of environmental calcium concentration on Ca²⁺ regulation has not been studied extensively in the primitive fishes, and may pose particular challenges to primitive cartilaginous fishes inhabiting fresh water, such as the acipenserids or sturgeons.

In fishes studied to date, the uptake of Ca²⁺ is achieved through the gills and the gastrointestinal tract (GIT) (Flik and Verbost, 1993), although the skin may play a role as well (Perry and Wood, 1985). The predominant uptake site is at the gills (Flik and Verbost, 1993; Flik et al., 1995), through mitochondria-rich cells and pavement cells (Shahsavarani and Perry, 2006), with the GIT playing a much smaller role (Bucking and Wood, 2007; Flik and Verbost, 1993; Sundell and Björnsson, 1988). In the gills, Ca²⁺ entry is a passive process, whereby Ca²⁺ enters epithelial cells via apical channels (Hwang and Lee, 2007), driven by very low intracellular concentrations (Flik, 1997). Once in the cell, Ca²⁺ is bound to proteins and actively pumped across the basolateral membrane into the blood by Ca²⁺-specific ATPases or exchanged for Na⁺ in Na⁺/Ca²⁺ exchangers facilitated by the Na⁺ gradient created by Na⁺,K⁺-ATPase (Hwang and Lee, 2007). In the GIT, Ca²⁺ uptake occurs following dietary ingestion (Bucking and Wood, 2007) and drinking (Flik et al., 1995; Guerreiro et al., 2004). Drinking may play a larger role in marine fishes, whereby the drinking of seawater offsets dehydrative water loss, and is also a potential source of Ca²⁺ uptake, particularly in juvenile fish (Flik et al., 1995; Guerreiro et al., 2002; Guerreiro et al., 2004). However, in hyperosmotic environments, GIT Ca²⁺ uptake may be minimized through reduced transcellular transport and increased excretion via paracellular pathways (Schoenmakers et al., 1993), and luminal Ca²⁺ obtained through drinking is largely precipitated by the bicarbonate base (Wilson et al., 2002). The uptake of Ca²⁺ in the GIT across the basolateral membrane of enterocytes relies more on Na⁺/Ca²⁺ exchangers than on Ca²⁺-ATPase (Flik et al., 1990; Schoenmakers et al., 1993).

To date, examination of calcium metabolism in fishes has almost been entirely devoted to teleost fishes. In freshwater cartilaginous fishes, because internal calcium stores are more limited than in teleosts, the GIT may play a greater role in Ca²⁺ uptake, supplementing gill uptake particularly during times of calcium need [i.e. juvenile growth, female vitellogenesis, low ambient calcium (Berg, 1968; Ichii and Mugiya, 1983)]. In addition, certain groups of freshwater cartilaginous fishes, such as the sturgeons, are known to consume calcium-rich invertebrate prey (Miller, 2004). Furthermore, because primitive fishes are closer phylogenetically to higher vertebrates, which rely on GIT uptake of Ca²⁺, mechanisms for Ca²⁺ uptake in primitive fishes may be more comparable to those used by higher vertebrates than to those of the more derived teleost fishes commonly studied.

In teleost fishes, increased calcium demands appear to be met through an upregulation of the epithelial Ca²⁺ channel (ECaC) in the gills (Liao et al., 2007; Shahsavarani and Perry, 2006). Interestingly, the active processes moving Ca²⁺ out of the cell at the basolateral membrane (i.e. Na⁺/Ca²⁺ exchangers and plasma membrane Ca²⁺-ATPases) do not appear to be upregulated (Liao et al., 2007). Endocrine control by the anti-hypercalcemic hormone stanniocalcin supports this, as it downregulates ECaC mRNA expression (Tseng et al., 2009). However, in some primitive fishes, such as sturgeons, the corpuscles of Stannius, the recognized source of stanniocalcin, are not present (Sasayama, 1999), indicating that the endocrine regulation of calcium may be somewhat different in primitive fishes. However, because studies in mammals have found that stanniocalcin is synthesized in a number of tissues rather than being released from an endocrine gland such as the corpuscles of Stannius, with similar regulatory effects in mammals as in teleosts (Gerritsen and Wagner, 2005; Hoenderop et al., 2005), the possibility does exist that stanniocalcin may be present in primitive fishes that lack corpuscles of Stannius.

ECaC has been characterized from a limited number of fishes (Pan et al., 2005; Qiu and Hogstrand, 2004; Shahsavarani et al., 2006), all of which are more derived bony fishes. ECaC has not been characterized in primitive fishes, nor have changes in mRNA levels been described, particularly in response to varying calcium concentrations typical of the aquatic environment. Furthermore, the role of ECaC has not been examined in the fish kidney in regard to environmental calcium concentrations. Although little is known in terms of kidney Ca²⁺-related function in primitive fishes (Wright, 2007), the kidney may play a large role in Ca²⁺ regulation owing to its importance for Ca²⁺ regulation in fishes (Björnsson and Nilsson, 1985) and higher vertebrates (Friedman and Gesek, 1995). Therefore, the objectives of this study were to characterize the function of ECaC in lake sturgeon, and to examine whole-body and intestinal Ca²⁺ flux as they relate to environmental calcium concentrations. Furthermore, the hypotheses that ECaC is upregulated in low calcium environments and that intestinal Ca²⁺ uptake plays a significant role in overall Ca²⁺ balance in this species were tested.

Juvenile lake sturgeon were derived from wild broodstock from the Winnipeg River, Manitoba, Canada, that were artificially induced to spawn. These juvenile fish were reared in 170 l tanks in dechlorinated tap water, and, for experimental purposes, were randomly divided into one of three calcium treatment groups, transferred to 120 l recirculating systems, and acclimated for at least 2 weeks in their respective treatments. The initial mean body mass of fish was not different between treatments (mean initial wet mass: 22.0±0.6, 23.0±0.7 or 23.7±0.5 g). CaCl₂ was used to adjust treatment calcium concentrations in recirculating systems to 0.1 (low), 0.4 (normal) or 3.3 (high) mmol l^–1 environmental calcium. The low concentration was arrived at following a pilot experiment in which fish health appeared compromised at a calcium concentration of 0.03 mmol l^–1 (i.e. red fin membranes, loss of equilibrium and death). Normal calcium water conditions were based on a 45 year dataset for water concentrations in the Winnipeg River system from sampling locations up to 50 km apart (Bing Chu, Environment Canada, personal communication), which showed relatively little variance (0.1 mmol l^–1) over time or location. The high calcium concentrations were higher than typical circulating concentrations in lake sturgeon (Allen et al., 2009c). The chemistry of treatment water was otherwise made to mimic natural conditions of the Winnipeg River through the addition of NaCl, KCl, MgSO₄ and Na₂HPO₄, and NaHCO₃ was used to adjust pH within a range of 7.3–7.6 (Table 1). Only Cl^– was unlike natural conditions owing to the unavoidable increase through addition of CaCl_2, but the use of CaCl₂ was preferable to alternatives, such as CaCO₃ (low solubility) or Ca(NO₃)₂ (potential toxic effects in a recirculating system). The use of CaCl₂ is also supported by previous studies examining environmental effects of calcium in teleost fishes (Flik et al., 1986; Hwang et al., 1996).

Fish were fed bloodworms twice daily at a total of 1.5% body weight/day based on dry mass of bloodworms. Calcium content of bloodworms was low (0.0079±0.0005 mmol l^–1 calcium g^–1 wet bloodworms, N=6) based on 2 mol l^–1 HNO₃ digestion. Fecal matter and uneaten food were removed 30 min after the addition of food. To maintain water quality, water was continually flushed through aquarium canister filters, and a 50% water change was conducted daily. Fish were maintained at 16°C in a temperature-controlled room, and kept on a 12 h light:12 h dark photoperiod during the experiment. Daily water samples were collected from each treatment throughout the acclimation period and analyzed by ion chromatography (Metrohm-Peak, Herisau, Switzerland) or atomic absorption spectrometry (AA240FS, Varian Inc., Palo Alto, CA, USA) for ion content. A different group of fish was used for each of the described experiments. All experiments were conducted under animal protocol F05-021, approved by the animal protocol management review committee of the University of Manitoba.

Following acclimation, fish were quickly transferred to individual flux chambers with a minimum body mass:volume ratio of 1 g:10 ml (Wood, 1992) covered with an opaque lid to reduce light intensity and potential effects of stress. The chambers contained 300 ml of water at the same ion concentrations and temperature as the treatment water. An airstone was placed in each chamber to maintain oxygen concentration and assist in mixing. All water, tissue digest and plasma samples were run in duplicate on a scintillation counter and atomic absorption unit. All experiments were conducted at the same time of day.

Blood and liver samples did not contain tracer, indicating that there had been no degradation and subsequent uptake of the radiolabel.

Degenerate primers were used to isolate fragments of the lake sturgeon ECaC sequence. An RNA ligase-mediated rapid amplification of cDNA ends (RLM RACE) commercial kit (Ambion, Inc., Austin, TX, USA) was used to obtain 5′ and 3′ ends of lake sturgeon ECaC. These were cloned using a commercial kit (Invitrogen, Carlsbad, CA, USA) inserted into a plasmid vector (Topoisomerase I from Vaccinia virus) and transformed by heat shock into chemically competent Escherichia coli (DH5α-T1), which were then plated onto LB agar plates treated with 50 μg ml^–1 ampicillin following the manufacturer's directions. Positive transformant clones were determined through X-gal staining and polymerase chain reaction (PCR) amplification with provided primers, subsequently amplified in 5 ml of LB medium containing 100 μg ml^–1 ampicillin, and then purified for DNA sequencing through centrifugation using a commercial kit (Miniprep, QIAGEN Inc., Mississauga, ON, Canada). Sequencing was conducted on a Hitachi 3130 Genetic Analyzer using ABDNA Sequencing Analyzing software (Applied Biosystems, Foster City, CA, USA).

Eight fish per treatment were euthanized and sampled for blood as described above, and 100 mg gill, kidney and GIT (pyloric caeca, mucosal scraping from mid-intestine, and spiral intestine) tissue samples were immediately removed, placed into 1 ml of RNAlater (Ambion), snap-frozen in liquid N₂ and subsequently stored at –80°C. Afterwards, blood plasma ion concentrations were analyzed using ion chromatography and atomic absorption spectrometry. Tissue samples were thawed, removed from RNAlater, and total RNA extracted using TRIzol reagent (Invitrogen). Manufacturer's instructions were followed and samples were resuspended in 40–200 μl of RNA storage solution (Ambion), and stored at –80°C.

Total RNA concentration was determined by absorbance at 260 nm and purity by absorbance at 260/280 nm. RNA was diluted in nuclease-free water to a concentration of 0.25 μg μl^–1, and the concentration was reconfirmed by spectrophotometer. RNA (0.5 μg) was treated with DNase (DNase 1, Invitrogen) according to the manufacturer's instructions, and tested for DNA contamination by PCR (40 cycles) using lake sturgeon-specific primers designed to amplify DNA coding for lake sturgeon EF-1α (EF-1α RTF1/RTR1, product size 237 bp; see Table 2). Complementary DNA (cDNA) was reverse transcribed from total RNA using oligo (dT) primers and Thermoscript reverse transcriptase (Invitrogen). The quality of the generated cDNAs from all tissues was assessed by employing the primer pair EF-1α RTF1/RTR1 in PCR, running the products on an electrophoresis gel, and visualizing with ethidium bromide and ultraviolet light. cDNA was diluted 1:2 in nuclease-free water and stored at –20°C.

Real-time PCR was conducted using a thermocycler (Miniopticon, Bio-Rad, Mississauga, Ontario, Canada) with iQS SYBR Green Supermix (Bio-Rad) in a 20 μl assay. For ECaC, 8 μl of diluted cDNA template were used, and 1 μl of each primer at a concentration of 10 μmol l^–1 were used. All primers were designed and optimized for the following PCR reaction conditions: an initial denaturing at 94°C for 4 min, followed by 45 cycles of 3 s at 94°C, 5 s at 64°C and 1 s at 72°C. Elongation factor 1α (EF-1α) was used as a reference gene based on its promise in other fishes (Tang et al., 2007). Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used as a reference gene for gills owing to the presence of treatment effects on EF-1α in gill tissue only. For EF-1α and GAPDH, 2 μl of diluted cDNA template were used. For EF-1α the PCR reaction conditions were: initial denaturing at 94°C for 4 min, followed by 40 cycles of 15 s at 94°C, 20 s at 64°C and 19 s at 72°C. For GAPDH the PCR reaction conditions were: initial denaturing at 94°C for 4 min, followed by 40 cycles of 15 s at 94°C, 20 s at 48°C and 30 s at 72°C. Following the PCR reactions, a melting curve from 40–94°C and subsequent gel verification were conducted to check for the presence of clean target products. Absolute ECaC, EF-1α and GAPDH mRNA levels were also quantified for tissues using carefully constructed standard curves derived from purified gel product (QIAquick gel extraction kit, QIAGEN Inc.), which were run in each real-time PCR assay for ECaC from 10^–4 to 10^–12 ng μl^–1, and for EF-1α and GAPDH from 10^–2 to 10^–8 ng μl^–1 (a total of N=6 for all tissues except kidney, where N=8). Primers for real-time PCR are listed in Table 2.

Statistical analysis was conducted using JMP (version 4.0.4; SAS Institute, Cary, NC, USA). Values presented are means ± s.e.m., and were compared using one-way analysis of variance (ANOVA) followed by Tukey's pairwise comparison. If values could not be transformed (log₁₀) to meet parametric assumptions of normality and homogeneity of variance, a non-parametric Kruskall–Wallis test was conducted followed by a Dunn's pairwise comparison. In all cases differences were considered significant at P<0.05.

Whole-body Ca²⁺ influx was significantly greater in the low calcium treatment group than in the high calcium treatment group (Fig. 1). Conversely, Ca²⁺ efflux was lower in the low calcium treatment group than in the high calcium treatment group, and was much lower than influx in all groups. Thus, the resulting net flux was strongly directed inward in all treatments, and increased in a stepwise progression with the lowest values of net flux in the low calcium treatment group and the highest in the high calcium treatment group.

In terms of plasma regulation, ionic or free Ca²⁺ was not different between treatment groups (Table 3). There were no differences between other ions as well, although total plasma ions for the low calcium treatment were significantly greater than for the high calcium treatment.

Total counts of ⁴⁵Ca from the GIT of influx fish indicated that there was a treatment Ca²⁺ effect (Fig. 2A). This was supported by the follow-up experiment examining intestinal Ca²⁺ flux, in which it was found that net flux was significantly greater in the low calcium treatment group than in the normal or high calcium treatment groups, with flux directed inward in the low calcium group and outward in the normal and high calcium groups (Fig. 2B).

Drinking rate was not significantly different between the treatment groups (P=0.07; Fig. 3).

The lake sturgeon ECaC sequence was obtained through sequencing and cloning steps. The whole gene consisted of 2831 nucleotides with a reading frame of 2049 nucleotides and 683 amino acids (Fig. 4), a predicted molecular mass of 78,917 Da and an isoelectric point at pH 6.994. An ankyrin repeat region, phosphorylation sites, six transmembrane regions and a putative ion pore region were identified in the sequence. A phylogenetic tree based on the amino acid sequence of the lake sturgeon ECaC sequence and other known ECaC sequences was obtained using the neighbor-joining method (Saitou and Nei, 1987). Lake sturgeon ECaC is phylogenetically more distantly related to other fishes studied to date, and fits well into its evolutionary place as a basal bony fish, between the African clawed frog (Xenopus laevis) and zebrafish (Danio rerio; Fig. 5). The lake sturgeon ECaC amino acid sequence shares 58% identity with zebrafish (Accession number AAQ89712.1.1), 63% identity with rainbow trout (Accession number NP_001117927.1), 66% identity with fugu (Takifugu rubripes, Accession number AAP46137.1), 64% identity with crayfish (Procambarus clarkii, Accession number AAR19087.1), and 52% identity with X. laevis CaT1 (BAC24123.1), based on comparisons with published sequences in GenBank.

ECaC mRNA levels were by far the greatest in the gills, measurable in the kidneys, and very low in the GIT tissues (Fig. 6). In the gills, ECaC mRNA levels were greater in high calcium conditions than in normal calcium conditions (Fig. 7). In the kidneys, ECaC mRNA levels were lower in low calcium conditions than in high calcium conditions (Fig. 8A). In the pyloric caeca (Fig. 8B), mid-intestine (Fig. 8C) and spiral intestine (Fig. 8D), mRNA levels were below the level of detectability in many samples, and there were no treatment effects.

In the present study, the influence of environmental calcium concentrations on calcium regulation in a freshwater cartilaginous fish and the relative involvement of the intestine and ECaC were examined. Two very interesting findings were noted. First, the surprising result of an increase in gill and kidney ECaC mRNA levels in high calcium conditions. Second, although the intestine had a relatively minor role in Ca²⁺ uptake, unidirectional influx of Ca²⁺ increased in low calcium conditions, suggesting the involvement of other transporters besides ECaC in the GIT. These findings are discussed below.

Fish must compensate for the effects of changing external calcium concentrations. Under low environmental calcium concentrations, whole body calcium decreases (Chou et al., 2002; Flik et al., 1986; Hwang et al., 1996). Therefore, teleost fish are known to compensate through increased Ca²⁺ uptake (Chou et al., 2002; Flik et al., 1986; Hwang et al., 1996) and increased circulating concentrations of Ca²⁺ (Flik et al., 1986). Increased Ca²⁺ uptake is primarily facilitated via branchial routes because of low levels of drinking in fresh water (Guerreiro et al., 2004). Internally, elevated circulating Ca²⁺ concentrations appear to be maintained through increased concentrations of protein-bound Ca²⁺, at least in non-vitellogenic lake sturgeon (Allen et al., 2009c). In contrast, under high environmental calcium concentrations, teleost fish may have higher whole-body calcium concentrations (Prodocimo et al., 2007), as external calcium concentration appears to be the main factor contributing to Ca²⁺ uptake in normal to high calcium environments (Guerreiro et al., 2004).

In this study, lake sturgeon responded to low environmental calcium concentrations by increasing Ca²⁺ uptake and decreasing efflux, whereas in high environmental calcium concentrations the opposite was true. Ca²⁺ influx rates in lake sturgeon were lower than those reported in freshwater-acclimated larval tilapia [200–250 nmol g^–1 h^–1 (Chou et al., 2002; Hwang et al., 1996)], comparable to those reported in juvenile tilapia [∼110 nmol h^–1 g^–1; Flik et al. (Flik et al., 1986); Flik et al. (Flik et al., 1985)], and higher than those reported from juvenile rainbow trout [30–60 nmol h^–1 g^–1 (Hogstrand et al., 1994; Niyogi and Wood, 2006)], killifish [50–60 nmol h^–1 g^–1 (Prodocimo et al., 2007)] and juvenile Adriatic sturgeon [∼12 nmol h^–1 g^–1 (Fuentes et al., 2007)]. Ca²⁺ efflux rates in lake sturgeon were similar or lower than those in larval tilapia [∼9–20 nmol h^–1 g^–1 (Chou et al., 2002)] and juvenile tilapia [∼15–35 nmol h^–1 g^–1 (Flik et al., 1986; Flik et al., 1985)], and higher than those in Adriatic sturgeon [∼1 nmol h^–1 g^–1 (Fuentes et al., 2007)]. In this study, it is likely that acclimation time also influenced the magnitude of Ca²⁺-influx, as influx rates in rainbow trout (Oncorhynchus mykiss) subjected to low calcium environments have been demonstrated to be initially high, and to gradually decrease with the duration of acclimation (Perry and Wood, 1985). Furthermore, although not measured in this study, influx in different calcium concentrations is likely to be affected by Ca²⁺ influx maximal velocity (J_max) and the Michaelis–Menton constant (K_m), which have been found to increase and decrease, respectively, in freshwater larval teleosts in low as compared with high environmental calcium concentrations (Chen et al., 2003).

In lake sturgeon, a 10-fold greater Ca²⁺ influx than efflux rate resulted in a strong inwardly directed net flux, regardless of environmental calcium concentration. This relationship has also been found in the developing stages of other fishes (Chou et al., 2002; Flik et al., 1986; Flik et al., 1985; Fuentes et al., 2007). Thus, lake sturgeon seem very capable of handling low calcium environments, at least at the concentrations tested, although there may be a lower limit of tolerable Ca²⁺ concentrations. In support of this, a pilot experiment found that fish lost equilibrium in calcium concentrations of 0.03 mmol l^–1.

An interesting finding of this study was the greater plasma ionic concentration in the low calcium treatment than the high calcium treatment (Table 3). In fishes, elevated environmental calcium is known to buffer against ion loss, and low environmental calcium and low pH can result in decreased osmolality and eventually decreased survival (Gonzalez et al., 2006; McDonald et al., 1980). However, the primary driver for ion loss is low pH rather than low environmental calcium, which causes Ca²⁺ efflux through paracellular tight junctions resulting in increased permeability of the gills (Gonzalez et al., 2006; Marshall, 1985). Thus, the ionic concentrations in the low calcium conditions in this experiment may be explained in part by the similar pH in all treatments. Furthermore, ion uptake capacity and affinity increases have been demonstrated in other fishes in ion-poor, low calcium conditions (Boisen et al., 2003). Thus, mechanisms for ion uptake may be enhanced under low calcium conditions. In addition, mechanisms for ion retention may be increased in ion-poor conditions, as increased bound Ca²⁺ has been found in wild, non-vitellogenic, female lake sturgeon in lower calcium environments (Allen et al., 2009c).

In most freshwater teleost fishes studied, Ca²⁺ uptake occurs predominantly at the gills. It was hypothesized that the intestine would have a larger role in Ca²⁺ uptake in lake sturgeon because of: the limited availability of internal concentrations of Ca²⁺ based on very low circulating concentrations of Ca²⁺ (Allen et al., 2009c); regressive scales with age (Peterson et al., 2007); the largely cartilaginous endoskeleton in which most Ca²⁺ is sequestered into structures such as the cranium (Findeis, 1997), which are less likely to be used for resorption purposes; and the known importance of the intestine to Ca²⁺ and base handling in sturgeons (Allen et al., 2009b). Thus, drinking and diet may play a larger role in Ca²⁺ uptake than is observed in teleost fishes.

In lake sturgeon, compensatory drinking to increase uptake of Ca²⁺ in low calcium conditions or decreased drinking in high calcium conditions does not appear to play a large role in Ca²⁺ homeostasis. Although the amount of ⁴⁵Ca in the GIT differed between treatment groups in the present study (Fig. 2A), drinking rates did not. In contrast, drinking rates have been shown to be responsive to environmental calcium concentrations in fishes, decreasing in high calcium concentrations in brown trout (Salmo trutta) (Oduleye, 1975). In lake sturgeon, drinking rates were comparable to those of other freshwater fishes in normal and high calcium environments, and were higher in low calcium environments, although much lower than those of typical seawater-acclimated teleosts (Hirano, 1974). Furthermore, GIT handling of Ca²⁺ and base, as precipitated CaCO₃ (Shehadeh and Gordon, 1969; Wilson et al., 2002), was not present in the different environmental calcium treatments of this experiment, although it has been documented in sturgeon acclimated to brackish water and seawater conditions (Allen et al., 2009b), where drinking plays a large role in preventing dehydration.

In lake sturgeon, net intestinal uptake of Ca²⁺ was occurring in low calcium environments, whereas net excretion was occurring in normal and high calcium environments. However, even in low calcium environments, the estimated proportion of intestinal Ca²⁺ uptake to whole-body Ca²⁺ uptake was small (≤2% either calculated as ⁴⁵Ca uptake in GIT compared with whole-body ⁴⁵Ca influx or calculated by drinking rate and water calcium concentration), with most uptake presumably occurring at the gills. Similar proportions of intestinal Ca²⁺ uptake have been found in other freshwater fishes (Flik et al., 1985), whereas in marine Atlantic cod (Gadus morhua), the intestine may contribute up to 30% of whole-body Ca²⁺ uptake (Sundell and Björnsson, 1988). When drinking rates are reduced by low salinities in marine larval gilthead seabream (Sparus auratus), contribution of the intestine to whole-body Ca²⁺ uptake greatly diminishes (<10%) (Guerreiro et al., 2004). Indeed, based on the drinking rates in this study, drinking does not appear to be a primary mechanism for compensatory Ca²⁺ uptake in lake sturgeon. Interestingly, the capacity for intestinal Ca²⁺ uptake based on the isolated flux studies (Fig. 2B) appears to be higher than that demonstrated by drinking rate. Therefore, the relative role of the intestine for Ca²⁺ uptake may be better elucidated through additional studies that evaluate dietary uptake as well. A number of studies have shown that intestinal uptake of Ca²⁺ may increase during sexual maturation or under low ambient calcium concentrations in fishes (Berg, 1968; Sundell and Björnsson, 1988; Ichii and Mugiya, 1983). Finally, although only the anterior-middle intestine was examined for Ca²⁺ flux in the present study, the stomach may also be a significant source of Ca²⁺ uptake (Bucking and Wood, 2007), as well as the spiral intestine.

ECaC is a member of the transient receptor potential (TRP) superfamily and the TRPV (vanilloid) subfamily (den Dekker et al., 2003). Although mammals are known to have two isoforms, ECaC1 (TRPV5) and ECaC2 (TRPV6) (den Dekker et al., 2003), only a single isoform has been identified in teleost fishes (Qiu and Hogstrand, 2004). The teleost fish ECaC shares characteristic structural features with the mammalian forms, having three ankyrin repeats, six transmembrane domains, and a putative hydrophobic Ca²⁺ pore between segments 5 and 6 (den Dekker et al., 2003; Perez et al., 2008; Qiu and Hogstrand, 2004). At the amino acid level, lake sturgeon ECaC is somewhat shorter than other fish ECaCs; however, the phylogenetic relationship of lake sturgeon ECaC fits into its evolutionary place as a basal bony fish, between amphibians and more derived fishes (Fig. 5).

Lake sturgeon ECaC mRNA levels were by far the highest in gill tissue as compared with in other tissues, which is consistent with other fishes studied to date (Pan et al., 2005; Qiu and Hogstrand, 2004; Shahsavarani et al., 2006). Current understanding of Ca²⁺ uptake in the gills of fishes involves transcellular transport via passive entry by apically located ECaC, and active transport at the basolateral membrane via plasma membrane Ca²⁺-ATPase (PMCA2) and Na⁺/Ca²⁺ exchangers (NCX1b) (Hwang and Lee, 2007), occurring in mitochondria-rich cells and pavement cells (Shahsavarani and Perry, 2006). Interestingly, in teleost fish ECaC appears to be regulated in response to environmental calcium concentrations (Craig et al., 2007; Liao et al., 2007; Pan et al., 2005; Shahsavarani and Perry, 2006), whereas PMCA and NCX are not (Liao et al., 2007). Indeed, the expression studies on teleost fish ECaC are analogous to those in the mammalian nephron, where ECaC has been described as the ‘gatekeeper’ for Ca²⁺ transport (Hoenderop et al., 2002). Contrary to this model, ECaC mRNA levels did not increase in low calcium conditions in lake sturgeon, but actually increased in high calcium conditions. Clearly, this finding is interesting, and may relate to several causes. First, ECaC expression has not been studied under high calcium conditions (greater than internal circulating Ca²⁺ concentrations) in other fishes, and so it is unknown how mRNA expression may change in these environments for other fish species. Second, it is possible that the localization of ECaC may change under high calcium conditions, its regulation may change, or that it has a different role in lake sturgeon. Possible indications of the latter may be the uncommon life history strategy of lake sturgeon, having a largely cartilaginous endoskeleton in freshwater, the apparent lack of corpuscles of Stannius (Sasayama, 1999) providing anti-hypercalcemic control (Tseng et al., 2009), and less rigidity for ionic Ca²⁺ regulation (Allen et al., 2009c). There is also the possibility that translational expression of protein, which was not measured in this study, may differ from transcriptional expression of mRNA, or that ECaC expression may change with acclimation time.

Previous studies on fishes have found very low activities of kidney ECaC and did not subsequently investigate changes associated with environmental calcium (Qiu and Hogstrand, 2004; Shahsavarani et al., 2006). In lake sturgeon, kidney ECaC mRNA levels were detectable and varied in response to environmental calcium concentrations. In mammals, ECaC1 (TRPV5) expression predominates in the kidney, where it is located on the apical surface of epithelial cells along the distal convoluted tubules, and it has a role in Ca²⁺ reabsorption (den Dekker et al., 2003). In the lake sturgeon, kidney ECaC expression decreased in low calcium conditions, which would seem to indicate a role in Ca²⁺ extrusion. When coupled with the environmental expression results from the gills, ECaC expression is the reverse to that previously described in teleost fishes. That is, ECaC expression increases in the gills in high calcium conditions when extrusion would be necessary, and decreases in the kidney in low calcium conditions when one would hypothesize that increased reabsorption would be required.

In lake sturgeon intestinal tissues (pyloric caeca, mid-intestine, and spiral intestine), ECaC mRNA levels were very low, and were below the level of detectability in many of the fish regardless of environmental calcium concentration. Intestinal ECaC, despite high expression levels in mammals (ECaC2, TRPV6) (den Dekker et al., 2003), has very low expression in teleost fishes (Qiu and Hogstrand, 2004; Shahsavarani et al., 2006), which is consistent with the findings in lake sturgeon in this study. Because lake sturgeon were found to regulate Ca²⁺ net flux in the intestine, it is unlikely that ECaC is the main intestinal Ca²⁺ transport mechanism in fishes. An L-type voltage-gated Ca²⁺ channel on the brush border of enterocytes is involved in Ca²⁺ uptake in Atlantic cod (Larsson et al., 1998). Ca²⁺ uptake could occur similar to the transcellular model proposed for the gill, or from a variety of processes, such as vesicular-mediated transport or paracellular pathways (Khanal and Nemere, 2008; Perez et al., 2008).

The freshwater, cartilaginous, lake sturgeon exhibits compensatory Ca²⁺ uptake in low calcium environments. Although net flux increases in the intestine in low calcium environments, drinking rates are low, and the proportional amount of intestinal Ca²⁺ influx to whole-body influx is low. ECaC plays a role in Ca²⁺ regulation in the lake sturgeon, with the gill being of primary importance, followed by the kidney, and very low expression in the intestine. ECaC expression is different from known models in freshwater fishes, with expression increasing in the gills in high calcium conditions and decreasing in the kidney in low calcium conditions. The uncommon life history strategy of a cartilaginous fish in freshwater, the phylogeny of this ancient fish, the low internal Ca²⁺ concentrations and the lower rigidity of Ca²⁺ regulation all indicate that different regulatory strategies may occur in the lake sturgeon. Future research investigating endocrine-mediated hypercalcemic control may be fruitful owing to the apparent lack of corpuscles of Stannius in sturgeon. Furthermore, localization of ECaC and investigations of other uptake mechanisms in the intestine, and possibly in the gills and kidneys, are needed to clarify potential differences in the Ca²⁺ transport mechanisms for this species.

We thank T. Allen for perspective and insight; two anonymous reviewers for providing constructive comments in the review of this manuscript; and T. Smith and the animal care staff at the University of Manitoba for assistance with fish care.